turboderp commited on
Commit
51b0469
·
1 Parent(s): 278786e

Delete README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -17
README.md DELETED
@@ -1,17 +0,0 @@
1
- EXL2 quants of Mistral-7B-instruct
2
-
3
- Converted from [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1). This is a
4
- straight conversion, but I have modified the config.json to make the default context size 7168 tokens, since in
5
- initial testing the model becomes unstable a while after that. It's possible that sliding window attention will
6
- allow the model to use its advertised 32k-token context, but this hasn't been tested yet.
7
-
8
- [2.50 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/2.5bpw)
9
- [2.70 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/2.7bpw)
10
- [3.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/3.0bpw)
11
- [3.50 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/3.5bpw)
12
- [4.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/4.0bpw)
13
- [4.65 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/4.65bpw)
14
- [5.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/5.0bpw)
15
- [6.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/tree/6.0bpw)
16
-
17
- [measurement.json](https://huggingface.co/turboderp/Mistral-7B-instruct-exl2/blob/main/measurement.json)